Hyperbolic Multimodal Representation Learning for Biological Taxonomies
This work addresses the problem of organizing biological specimens into hierarchies for biodiversity researchers, but it is incremental as it builds on existing hyperbolic and multimodal methods.
The paper tackled taxonomic classification in biodiversity research by investigating hyperbolic networks as an embedding space for hierarchical models, achieving competitive performance with Euclidean baselines and outperforming other models on unseen species classification using DNA barcodes.
Taxonomic classification in biodiversity research involves organizing biological specimens into structured hierarchies based on evidence, which can come from multiple modalities such as images and genetic information. We investigate whether hyperbolic networks can provide a better embedding space for such hierarchical models. Our method embeds multimodal inputs into a shared hyperbolic space using contrastive and a novel stacked entailment-based objective. Experiments on the BIOSCAN-1M dataset show that hyperbolic embedding achieves competitive performance with Euclidean baselines, and outperforms all other models on unseen species classification using DNA barcodes. However, fine-grained classification and open-world generalization remain challenging. Our framework offers a structure-aware foundation for biodiversity modelling, with potential applications to species discovery, ecological monitoring, and conservation efforts.